Built using Python, dxlib focuses on providing a comprehensive set of tools for quantitative finance.
The library is designed to be easily adapted and used with other libraries such as pandas and numpy, and inserted into existing workflows.
For low-level development and HFT, look at the companion library dxcore. dxlib is an interface to dxcore, with additional API and network interfacing, QPM, and other tools.
Seeing as some existing libraries targeting quantitative development already exist, the goal of dxlib is to provide easier and act as a more maintainable library for both big and small projects.
- QuandL has been archived, and dxlib is a great alternative.
- QuantLib is a great library, and should be used in conjunction with dxlib, but its focus differs.
Eventually,
dxcoreshould replaceQuantLibwithin the context of dxlib. - pandas and numpy are great, and should be used in conjunction with dxlib.
All modules and classes are built using Domain Driven Design, and are designed to be easily understood and used. I myself come from a computer science background, and whenever starting a new quant project, always found my code to end up extremely convoluted and messy. Therefore, I believe creating a library with a strong foundation rather than a collection of scripts is the way to go.
All classes and methods are supposed to be easily serializable, deserializable, and extendable - to be freely used in a distributed and/or parallel environment. For now, the cache system uses both HDF5 and JSON, and the networking system allow for easily interfacing with other systems. Current inbuilt handlers include REST and Websockets. Future encodings are planned to include FIX, SBE. Future handlers are planned to include ZeroMQ, gRPC and rough UDP.
In the future, dxlib will be able to interface with dxcore for low-level development, and dxforge for high-level network development. Look at dxhft for implementation-specific high-frequency trading tools.